Guide · billing-unit dictionary
Tasks vs credits vs executions — what each word counts
A dictionary of three billing words. Totals are not published yet. Self-hosted is a separate lane. A score is not a monthly bill. This page does not add new scores.
Use this page if the three words do not match
Use
Use this page if…
You need the plain meaning of Zapier tasks, Make credits, and n8n executions: what each word counts, what it does not count, and why 1 is not 1 across the three tools. Stay on the words. Dated quotes below were read 2026-09-30.
Skip
Skip if…
You want one job counted at 50 / 500 / 5,000 runs — Same job, three meters. You want loops, retries, and polls — How to meter automation. You are near a cap — When you hit the meter. You added an AI step — How AI steps change the meter. You are still on free or a trial — When free isn’t enough. You want a cheapest-tool list. This page does not rank products.
A use-or-skip check for the three billing words.
- You need the dictionary: what a Zapier task, a Make credit, and an n8n execution count.
- You will stay on the words — what each one counts, and what it does not.
Read more
What this page tells you. What the word task counts on Zapier, what the word credit counts on Make, and what the word execution counts on n8n — and what each word leaves out. Companion to the 5-step job, the watch list, the ceiling, the AI delta, and the free-plan read. This page shows units and definitions only, with dated quotes from primary vendor pages, accessed 2026-09-30. It does not invent a monthly dollar, a converted list, or a savings figure. This page publishes no annual savings. Verify before buy. It does not add new scores.
GuideSome scores not shown yetVendor pages accessed 2026-09-30Live scorecard
This page vs the other meter reads
Same job, three meters counts one 5-step job as tasks, credits, and executions at 50 / 500 / 5,000 runs. This page does not reprint that volume ladder.
How to meter automation is what to watch — usage, volume, and credits — and what multiplies a job. This page does not reprint that watch list.
When you hit the meter is the ceiling: hold, extra credits, fail, or upgrade. This page does not reprint that ceiling.
How AI steps change the meter is the AI / agent delta. This page names that some rates differ, then stops. It does not reprint those formulas.
When free isn’t enough is the free or trial shape versus the first paid plan. This page does not reprint those plan gates.
This page is the dictionary. Three words. What each one counts. What it does not. Why 1 is not 1. Units and definitions only. It does not invent a monthly dollar.
The three words
Zapier’s word is a successful action. Make’s word is, by default, a module run. n8n’s word is the whole production run. Dated quotes below are from primary vendor pages, accessed 2026-09-30.
Task
Zapier
A task is any successful action that runs in Zapier. Only successful actions count toward your task usage.
The same page lists what counts and what does not. A trigger is not a task. A successful action is. Some Zapier products use a different task rate than one per action. This dictionary names those exceptions. It does not reprint the AI-step formula.
Credit
Make
Credits are the currency you buy and consume to use Make. Only features triggered by scenario runs (apps, modules, and some in-app features) or the AI agent’s chat use credits.
Make’s credits page says credits replaced operations as the term for the billing unit. By default, 1 operation equals 1 credit. An operation is a single module run to process data or check for new data. A trigger module runs once per check.
Execution
n8n
An execution is a single run of your entire workflow. It doesn’t matter how many steps are in the workflow or how much data it processes. It’s still a single execution.
The executions docs use the shorter line: an execution is a single run of a workflow. Paid-plan quota counts production executions. Manual tests, sub-workflow runs, and error-workflow runs do not add a quota execution.
What each word counts, and what it does not
Read down one column at a time. The “also” line is a published exception to the base word, still in units. A blank on a verified page stays —.
Swipe sideways on a narrow screen. Not a scored table.
| Word | Counts | Does not count | When the word is not 1 |
|---|---|---|---|
| Zapier task | One successful action step. Also a successful step on an error-handler path, a previously successful step that runs again on a full Zap replay, each action inside a sub-Zap, Call a Sub-Zap, Return from a Sub-Zap, and a search set to proceed if nothing is found. | All trigger steps, including polls. Filter or Paths. Steps that error, halt, or do not run. The Start a Sub-Zap trigger. Zapier Tables and Zapier Forms steps. Formatter, Delay, Looping, the Sub-Zap app, Digest, Zapier Manager, and Storage. Zapier Agents do not count as tasks. | The base word is one successful action. Zapier also publishes other rates on the same page: Code by Zapier is 1 task per run at standard runtime and can use more if you extend the runtime; Lead Router is 5 tasks per successful lead routed; MCP is 2 tasks per successful tool call. AI by Zapier depends on the model tier — that formula is on How AI steps change the meter. |
| Make credit | By default, one operation. An operation is one module run that processes data or checks for new data. A trigger check is one operation, however many bundles it returns. Later modules run once per bundle. Make’s credits page says only scenario-run features and the AI agent’s chat use credits. | A router does not consume credits. When an error handler activates, it does not consume operations. A filter’s credit cost is not stated on the credits, operations, router, or error-handling pages verified for this dictionary: —. | The default word is 1 operation equals 1 credit, and that rate does not change with file size. Some AI and advanced modules use a varying number of credits. That token table is on How AI steps change the meter. This page does not reprint it. |
| n8n execution | One production run of the whole workflow. A schedule trigger counts one execution every time it fires. A polling trigger counts one execution only when it finds new data. A webhook counts one execution for every inbound request that starts the workflow, including an empty body. | Manual runs from the editor. A sub-workflow called with Execute Sub-workflow — only the parent execution counts. A workflow set as an error workflow. A poll that returns no data. A malformed webhook request that fails before the workflow starts. | Step count does not change the word. n8n’s pricing FAQ says it is still a single execution however many steps the workflow has. Gateway credits are a different balance. That read is on How AI steps change the meter. |
Zapier’s task page, accessed 2026-09-30, is the source for the task column. Make’s credits page, operations page, router page, and error-handling overview are the source for the credit column. n8n’s executions docs and pricing FAQ are the source for the execution column. Router line, as published: routers don’t consume credits. Error-handler line, as published: when an error handler activates, it doesn’t consume operations.
Why 1 is not 1
The same moment in a job is a different object on each meter. This table is the rule. It is not a 5-step inventory and it is not a volume ladder.
Swipe sideways on a narrow screen. Not a scored table.
| Object | Zapier task | Make credit | n8n execution |
|---|---|---|---|
| The check that starts the job | Does not count. Trigger steps are outside task usage. Polls never use tasks. | Counts. A trigger module is one operation per check. By default that is one credit. | Sits inside one production execution. It is not a separate execution. |
| One successful action in that run | Counts as one task. The word is a successful action. | Counts as one operation, and by default one credit. The word is a module run. | Still that same execution. A step is not an execution. |
| A second successful action in the same run | Counts as another task. | Counts as another operation, and by default another credit. | Still one execution. The whole workflow is the unit. |
One task is not one credit is not one execution. The trigger is the clearest split: 0 Zapier tasks, 1 Make operation (and by default 1 credit), and still inside 1 n8n execution. A second successful action adds a task and a credit. It does not add an execution. Verify before buy.
Don’t turn a word into a bill
A task, a credit, and an execution are units. A /500 score is not a plan. This page does not invent a converted list. It publishes no annual savings.
- Stay on the word until you can say what it counts and what it does not.
- Then open the worked example on Same job, three meters if you need the same job at 50, 500, and 5,000 runs.
- Leave the ceiling, the AI step, and the free plan on their own pages. This dictionary does not invent the dollar between the words.
Units and definitions only. Check the vendor page on the day you buy.
How this sits on the scorecard
How this sits on the scorecard
Totals are not published yet.
If you still need to pick a product before you use the words, see Which automation tool for small teams. When-to-pick copy lives on Make vs n8n vs Zapier — when to pick each.
Product links
Sources
Primary vendor pages. Access date for every unit definition on this page: 2026-09-30. List prices on the vendor pricing pages are omitted. We do not invent search volumes, testimonials, or a converted list.
- Zapier — how task usage is measured · accessed 2026-09-30
- Make — credits · accessed 2026-09-30
- Make — operations · accessed 2026-09-30
- Make — overview of error handling · accessed 2026-09-30
- Make — add a router · accessed 2026-09-30
- n8n — understand executions · accessed 2026-09-30
- n8n pricing · accessed 2026-09-30
FAQ
What does a Zapier task count?
A task is any successful action that runs in Zapier. Only successful actions count. Triggers, Filter, Paths, Formatter, Delay, Looping, Digest, Zapier Manager, Storage, the Sub-Zap app, Zapier Tables, and Zapier Forms do not count. Steps that error, halt, or do not run do not count. Zapier Agents do not count as tasks. Code by Zapier is 1 task per run at standard runtime. Lead Router is 5 tasks per successful lead. MCP is 2 tasks per successful tool call. AI by Zapier uses a model-tier rate — that formula is on How AI steps change the meter. Accessed 2026-09-30.
What does a Make credit count?
Credits replaced operations as Make’s billing-unit term. Credits are what you buy and consume. Only features triggered by scenario runs, or the AI agent’s chat, use credits. By default, 1 operation equals 1 credit. An operation is one module run that processes data or checks for new data. A trigger check is one operation. Routers do not consume credits. When an error handler activates, it does not consume operations. A filter’s credit cost on the pages verified for this dictionary: —. Dynamic AI credit rates are on How AI steps change the meter. Accessed 2026-09-30.
What does an n8n execution count?
An execution is a single run of a workflow. n8n’s pricing FAQ says it is a single run of the entire workflow, however many steps it has and however much data it processes. Paid-plan quota counts production executions only. Manual runs, sub-workflow runs, error-workflow runs, empty polls, and webhook requests that fail before the workflow starts do not count. Accessed 2026-09-30.
Why is 1 not 1 across Zapier, Make, and n8n?
The three words count different objects. A Zapier task is a successful action, and the trigger is not one. A Make credit is, by default, one module run, and the trigger check is one. An n8n execution is the whole production run, so a second action does not add a second execution. One task is not one credit is not one execution. This page does not multiply that rule by a monthly volume.
Does this page turn those words into a monthly dollar?
No. This page stays on the words: what each one counts and what it does not, as printed on primary vendor pages, accessed 2026-09-30. It does not invent a monthly dollar, an overage dollar, or annual savings. It publishes no annual savings. The 5-step volume ladder is on Same job, three meters. The ceiling is on When you hit the meter.
Does this guide change the Make vs n8n vs Zapier scores?
No. Totals are not published yet.
Where should I go next?
To count one 5-step job at 50, 500, and 5,000 runs, see Same job, three meters. For loops, retries, and polls, see How to meter automation. For an AI or agent step, see How AI steps change the meter. If you are near a cap, see When you hit the meter. If you are still on free or a trial, see When free isn’t enough. To pick a product, see Which automation tool for small teams. Totals are not published yet.
Affiliate disclosure
Make Partner (pc=stackscorecard) is live elsewhere on the site and is not linked from this page. n8n and Zapier stay official-site URLs with UTM only — no invented IDs — and rankings are not sold (Affiliate disclosure · Disclosures).